Summary counts of comments on issues by users in a github repo

get_issue_comments.py

import requests
import pandas as pd
import altair as alt
from typing import List, Tuple


def download_comments(repo: str, user: str) -> List[str]:
    url = f"https://api.github.com/repos/{user}/{repo}/issues/comments"
    res = requests.get(url)
    if res.ok:
        return res.json()
    else:
        raise Exception("Couldn't download the data from github")

def get_dates_and_users(comments: List[str] = None) -> pd.DataFrame:
    """Parse the output from download_comments()"""
    
    ret: List[Tuple[str, str]] = []

    for comment in comments:
        ret.append((comment["created_at"], comment["user"]["login"]))
    
    df = pd.DataFrame(ret, columns=["date","user"])
    
    # set the date as the index
    df['date'] = pd.to_datetime(df['date']) 
    df = df.set_index('date')
    return df


def get_summary_df(df, timeframe: str = "week") -> pd.DataFrame:
    if timeframe == "week":
        return df.groupby("user").resample('W').count().rename({"user": "count"}, axis=1).reset_index()                                                                                    
    elif timeframe == "month":
        return df.groupby("user").resample('M').count().rename({"user": "count"}, axis=1).reset_index()                                                                                             
    else:
        print(f"Using timeframe of {timeframe} is not supported.")
        raise

repo = ""
user = ""
comments = download_comments(repo, user)
df = get_dates_and_users(comments)
wrangled = get_summary_df(df)
chart = (alt.Chart(wrangled)
    .mark_line(point=True)
    .encode(x="date",
            y="count",
            color="user",
            tooltip=['user', 'count']
    )
    .properties(
        title=f"Comments on issues in {repo} over time"
        )
    .interactive()
    )
chart.save(f"{repo}_comments.html")